Field of the Invention
This invention relates to systems and methods for auditing and correcting data inconsistencies in data replication systems.
Background of the Invention
In z/OS Global Mirror, data is asynchronously mirrored from a primary storage system to a secondary storage system in order to maintain two consistent copies of data. The primary and secondary storage systems may be located at different sites, perhaps hundreds or thousands of miles away from one another. In the event the primary storage system fails, I/O may be redirected to the secondary storage system, thereby enabling continuous operations.
When updates are received at the primary storage system, z/OS Global Mirror groups the updates into record sets and collects these record sets into data structures referred to as “sidefiles.” These sidefiles are stored in cache at the primary storage system. The record sets in these sidefiles are then asynchronously mirrored to the secondary storage system using data replication software (also known as System Data Mover, or SDM, in the z/OS environment) which is typically (but not necessarily) hosted on a secondary host system.
Currently, when asynchronously mirroring data from a source volume to a target volume, it is very difficult to determine if there are inconsistencies in the mirrored data. When an error such as a No Record Found (NRF) or Invalid Track Format (ITF) error occurs at the target volume, the mirroring relationship is typically suspended and data in the source volume is recopied to the target volume. The mirroring relationship between the source volume and target volume may then be reestablished. If an error is detected long after an out-of-sync condition occurs, it may be difficult to diagnose the cause of the error. Currently, the only option to validate the consistency of mirrored data is to compare, track by track, the source and target volumes. This creates significant overhead and may result in false errors when updates to the target volume are in flight.
In view of the foregoing, what are needed are systems and methods to more effectively audit data consistency across source and target volumes in asynchronous data replication environments. Further needed are systems and methods to more efficiently correct errors in consistency when discovered.